Charles Sun

LinkedIn — © 2026 Charles Sun. All rights reserved. This article explains why the future of artificial intelligence cannot scale, secure itself, or operate coherently on an Internet still dependent on IPv4. Charles Sun argues that IPv6 is not an upgrade but the foundational requirement for AI‑era infrastructure — enabling end‑to‑end identity, global addressability, reduced attack surface, and the elimination of brittle dual‑stack complexity. Drawing on his role in shaping the U.S. federal IPv6‑only mandate, Sun outlines why sovereign AI, hyperscale compute, and next‑generation digital systems all require a clean, IPv6‑only architecture to function reliably and securely.

LinkedIn — © 2026 Charles Sun. All rights reserved. This article reveals a surprising security lapse: the world’s most visited website still negotiates TLS 1.0 and 1.1 in 2026. Charles Sun argues that deprecated cryptographic protocols persist not because they are needed, but because organizations fail to retire them. He makes the case that modern security posture requires using TLS 1.3 only — even though TLS 1.2 remains common and has not been formally deprecated — because older versions expand the attack surface and weaken cipher strength. Sun connects this to a broader pattern in Internet infrastructure, explaining why disciplined protocol retirement and IPv6-only deployment are essential to reducing systemic risk at global scale.

Homeland Security Today — A landmark reflection by Charles Sun, a federal IPv6 thought leader and cybersecurity visionary, this essay challenges conventional boundaries in digital evolution—tracing the strategic transformation from legacy internet foundations to next-generation, AI-driven, secure connectivity. Anchored in decades of public-sector innovation and mission-built architecture, it calls for renewed integrity, foresight, and global collaboration to safeguard the future of the connected world.

blogspot.com — In “If Not Now, When?: The Federal IPv6-Only Mandate After Five Years,” Charles Sun issues a clarion call: after years of delays, the time has come for the U.S. federal government to embrace 100 % IPv6-only networks—not as a luxury, but as a strategic security imperative. He argues that dual-stack IPv4/IPv6 operations inherently double the attack surface, complicate Zero Trust implementation, and undermine modernization goals. Drawing on policy milestones (notably OMB’s M-21-07), his own long advocacy, and the stark mathematics of address space, he dissects the technical, organizational, vendor, cultural, and accountability failures that have derailed progress. Then he pivots to a bold, actionable roadmap: renewed transparency, binding interim targets through 2028, a CISA-led Center of Excellence, procurement mandates for IPv6 parity, workforce investment, and leadership accountability. At its core, the essay insists: the mandate is immutable, the urgency is real, and the mission must now be completed—with resolve, discipline, and strategic clarity.

blogspot.com — This page serves as the central record of the IEEE plagiarism case. It documents the original 2020 exposure, the 2025 escalation after institutional inaction, and the supporting forensic evidence. The purpose is simple: to preserve the facts, ensure accountability, and provide a transparent record for colleagues, editors, and the global research community.

LinkedIn — © 2025 Charles Sun. All rights reserved. Five years after the first exposure, and days after a formal complaint deadline, IEEE remains silent. The matter now moves to escalation. This post is a sequel to my 2020 exposé on plagiarism in an IEEE publication. Five years after that initial documentation, I submitted a formal complaint directly to IEEE. The deadline I set for corrective action has now passed without response. In light of this continued silence, the matter must be escalated.

blogspot.com — Executive Summary This forensic analysis identifies multiple, verifiable instances of plagiarism within a July 2017 IEEE IoT Newsletter article. The investigation reveals that the piece systematically appropriates specific phrasing, unique statistical data, core argumentative structures, and central conceptual frameworks from Mr. Sun’s earlier work with no attribution. The types of plagiarism identified range from verbatim copying and patchwriting to the severe misappropriation of a unique intellectual concept and the intentional obfuscation of sources to disguise stolen research. These practices constitute a severe and multi-layered breach of academic and journalistic integrity, misrepresenting the originality of the work and failing to credit the original author for his research, synthesis, and innovative ideas.

LinkedIn — © 2025 Charles Sun. All rights reserved. This companion piece to “Bias Is Not a Bug” dives deeper into the neurocognitive roots of algorithmic bias. It explains how unconscious human judgments—formed through fast, automatic “System 1” thinking—shape AI long before model training begins. From biased data labeling to flawed aggregation, the article shows how implicit bias enters upstream and silently distorts outcomes. Drawing on research from Greenwald, Kahneman, Forscher, and others, it highlights the limits of one-off bias trainings and calls for structural safeguards: calibrated annotator pools, bias-aware aggregation, fairness gates in CI/CD, and lifecycle documentation. The message is clear: we don’t need perfect minds—we need engineered workflows that protect fairness from the start.

LinkedIn — © 2025 Charles Sun. All rights reserved. This article challenges the myth of AI neutrality by revealing how unconscious human biases—embedded in data, labels, and design choices—shape algorithmic behavior. From hiring tools penalizing gendered language to facial recognition misclassifying darker-skinned women, the piece shows how AI systems amplify societal inequities rather than correct them. Drawing on research from Buolamwini, O’Neil, and Kahneman, it explores the cognitive roots of bias and the illusion of objectivity in machine learning. The article calls for upstream safeguards: diverse annotators, bias-aware aggregation, and ethical oversight. AI doesn’t transcend human flaws—it mirrors them. And if we want better outcomes, we must start with better inputs.

LinkedIn — © 2025 Charles Sun. All rights reserved. AI hallucinations aren’t the real problem—it’s our willingness to trust them without question. Throughout history, misinformation has spread not because of technology, but because of human minds eager to believe. In this piece, I dive into why AI-generated errors aren't as "new" as they seem and how critical thinking remains the ultimate safeguard against blind certainty. Let’s challenge assumptions and rethink what it means to be wrong. Read more & join the conversation!

LinkedIn — © 2024 Charles Sun. All rights reserved. Cloud computing has revolutionized how organizations manage and process data, offering unparalleled scalability and flexibility. However, ensuring data security remains a significant challenge. Automated Machine Learning (AutoML) emerges as a promising solution to address both scalability and security concerns. This article examines the ways in which AutoML can be leveraged to enhance data security while achieving seamless scalability in cloud computing environments.

LinkedIn — © 2023 Charles Sun. All rights reserved. This piece explores the transformative potential of Automated Machine Learning (AutoML) in addressing both security and scalability challenges in cloud environments. Key Highlights: · Layered Security: How AutoML enhances network security, data encryption, and IAM. · Scalable Infrastructure: Leveraging AutoML for efficient resource management and microservices architecture. · Continuous Monitoring & Incident Response: Utilizing AutoML for real-time threat detection and automated response. · Compliance & Best Practices: Ensuring regulatory compliance with AutoML-driven audits and assessments. · Backup & Disaster Recovery: Optimizing data backup and disaster recovery processes with AutoML. AutoML provides a robust solution for securing and scaling cloud infrastructure, paving the way for innovative and resilient IT systems. Whether you're in IT, cybersecurity, or cloud services, I believe you'll find valuable insights in this article.

LinkedIn — © 2023 Charles Sun. All rights reserved. Charles Sun’s article responds to the notion that humans are simply an “API” for ChatGPT, a metaphor that’s occasionally floated in tech circles. Sun strongly rejects this analogy, arguing that humans are far more than just interfaces for AI: humans train, supervise, and ultimately control the development and deployment of AI systems like ChatGPT. He points out that if humans were only considered “APIs,” it would erase the critical role of human oversight, judgment, and common sense in guiding and restraining AI technology. Sun invokes examples from both AI and IoT to illustrate that human decision-making remains essential and irreplaceable in any AI development and application.

Homeland Security Today — As the next-generation Internet Protocol, IPv6 (Internet Protocol version 6) is the prerequisite for the future growth and development of the Internet. In the era of artificial intelligence (AI), 5G, and Internet of Things (IoT), there are increasingly more and more innovations and emerging technologies that will be heavily relying on the support of the Internet, which is currently running in the so-called dual-stack of mode operations using both Internet Protocol version 4 (IPv4) and Internet Protocol version 6 (IPv6). Moreover, adopting and enforcing the IPv6-only policy worldwide by securely deploying the single stack of IPv6, turning off IPv4, and setting a specific deadline to sunset IPv4 completely will dramatically reduce the overall cybersecurity threats and attacks based on IPv4.

Homeland Security Today — The ever-increasing daily cybersecurity incidents and confirmed data breaches across the globe have touched every aspect of our society and negatively impacted our daily lives.

Homeland Security Today — The future of the Internet depends on the continued growth of a solid, healthy, and secured underlying global network infrastructure supporting the demand for the next generation of the Internet using IPv6 (Internet Protocol version 6) as its communication protocol. The future of the Internet is depending on the global successful adoption of IPv6.

Homeland Security Today — In the era of the Internet of Things (IoT) and blockchain, it is a daily nightmare encountered by organizations from both public and private sectors, big and small, struggling to react to the aftermath of the constant cybersecurity breaches and ransomware attacks.

ComputerWorld — Do you think the Internet of Things (IoT) will be the Next Big Thing? It can't be. Not until we get past the real Next Big Thing: IPv6. Without the extensive global adoption and successful deployment of IPv6 as the primary version of the Internet Protocol, the IoT won't be possible.

ComputerWorld — Internet Protocol Version 6 (IPv6) is here for real. The U.S. Office of Management and Budget has set a deadline of Sept. 30 for government agencies to give employees the capability to access the Internet via IPv6. In fact, now is the time for all of us to start thinking about when we should stop using IPv4.